The Salience Developer Hub

Welcome to the Salience developer hub. You'll find comprehensive guides and documentation to help you start working with Salience as quickly as possible, as well as support if you get stuck. Let's jump right in!

Get Started

Salience 6.5.0 Release 2019-10-14 (r1741)

This is the Salience 6.5 release, introducing support for a greater variety of Assembler machine learning models, the ability to verify that models behave correctly and more including faster runtime speeds.


  • You can now specify a path for Salience to log information to when constructing a session.
  • Salience was leaving two environment variables behind when you ran the uninstaller on Windows.
  • Added option to specify a URL where the Basic Language Server is running if you want to split that across machines. This requires you to manage the server (launch it, restart it if it dies).
  • "Next of" rules in an entity relationship can now filter by entity labels as well as types.
  • Added DisableChunkParser option. This degrades accuracy for a number of features (sentiment, entities, ...) but speeds processing.
  • Themes/normalizations.dat no longer requires that you enter stemmed forms of themes. Fixed stemming of single word themes (don't exist by default but can be enabled via rules.ptn).
  • Relationships now return the specific mentions of the entity they matched, instead of all mentions of those entities.
  • More features supported inside custom models.
  • Added clearer labels for hierarchical model-based entity labels (e.g. Organization/Company/Bank).
  • Queries against a token can reference only mentions with a specific POS tag using the "_" character, for example: cook_VB will only match "cook" being used as a verb.
  • Performance enhancements to running custom models.

Bug fixes

  • Fixed bug where gluepatterns was appending a space on the start of the glued token.
  • Fixed bug when using the *{min,max} pattern syntax in a relationship file (since that also uses curly braces).
  • Fixed character offset errors near Chinese punctuation, and when using gluepatterns.dat to combine tokens.

Migration strategy

  • There are no API changes required to use Salience 6.5.
  • If you have fr-data-2485, which contains an old Assembler model, you must upgrade to fr-data-2781 or later if you are using Salience 6.5
  • If you have fr-data-2111, you can use this with Salience 6.5 and see the same results as before.
  • Ports 15980-16108 should not be firewalled on any machine running Salience 6.5 because they are now used for MLHandler and MLWorker processes.

If you have any questions, please email [email protected]

Updated about a year ago

Salience 6.5.0 Release 2019-10-14 (r1741)

Suggested Edits are limited on API Reference Pages

You can only suggest edits to Markdown body content, but not to the API spec.